Classical dendritic cells contribute to hypoxia-induced pulmonary hypertension

Abstract

Pulmonary hypertension (PH) is a chronic and progressive disease with significant morbidity and mortality. It is characterized by remodeled pulmonary vessels associated with perivascular and intravascular accumulation of inflammatory cells. Although there is compelling evidence that bone marrow-derived cells, such as macrophages and T cells, cluster in the vicinity of pulmonary vascular lesions in humans and contribute to PH development in different animal models, the role of dendritic cells in PH is less clear. Dendritic cells' involvement in PH is likely since they are responsible for coordinating innate and adaptive immune responses. We hypothesized that dendritic cells drive hypoxic PH. We demonstrate that a classical dendritic cell (cDC) subset (cDC2) is increased and activated in wild-type mouse lungs after hypoxia exposure. We observe significant protection after the depletion of cDCs in ZBTB46 DTR chimera mice before hypoxia exposure and after established hypoxic PH. In addition, we find that cDC depletion is associated with a reduced number of two macrophage subsets in the lung (FolR2⁺ MHCII⁺ CCR2⁺ and FolR2⁺ MHCII⁺ CCR2^-). We found that depleting cDC2s, but not cDC1s, was protective against hypoxic PH. Finally, proof-of-concept studies in human lungs show increased perivascular cDC2s in patients with Idiopathic Pulmonary Arterial Hypertension (IPAH). Our data points to an essential role of cDCs, particularly cDC2s, in the pathophysiology of experimental PH.

Keywords: cDC2; classical dendritic cells; hypoxia; inflammation; pulmonary hypertension.

FIGURE 1

cDC2s and their subset CD301b⁺ are increased in lungs of WT mice after 3 days of hypoxia exposure. Not much difference was observed in cDC1s under the same conditions. DCs CD11b⁺ and CD11b⁻ in the blood were slightly decreased, but the difference was not statistically significant. Panel A shows gating strategy for lung DCs. Singlets were initially gated as well as viable CD45⁺ cells (negative for viability dye (VD) and dump gate (VD + dump); parenchymal cells were then isolated (IV-CD45⁻), followed by gating on CD64⁻ cells. To remove eosinophils as well as any alveolar macrophages, we gated on Siglec-F negative populations, followed by DCs gating (CD11c⁺ and MHCII⁺). To distinguish cDC1s from cDC2s we gated the cells for CD103 and CD11b. cDC2s subsets were separated using CD301b. Panel B shows the absolute numbers of cDC2s, cDC2s CD301b⁺ and cDC1s. Panel C shows CD11b⁺ and CD11b⁻ DCs in the blood of the same animals. p Value * =.01 to .05, ** =.001 to .01, ns, not significant.

FIGURE 2

cDC2s, particularly their CD301b subset are activated after hypoxia exposure. Panels A and B are absolute numbers of all activated cells (CD80⁺, CD86⁺ or CD80⁺ and CD86⁺) of cDC2s and cDC1s respectively. Panels A1 and A2 are expression levels (Media fluorescence intensity or MFI) of CD86 and CD80 in cDC2s and its subset CD301b. Panel B1 shows expression levels CD80⁺ and CD86⁺ in cDC1s (as described in panel A). p value * =.01 to .05, ** =.001 to .01, *** =.0001 to .001, ns, not significant.

FIGURE 3

Classical dendritic cells (cDCs, ZBTB46⁺ in brown) are increased in the lungs of wild-type mice challenged with 3 days of hypoxia, particularly in the perivascular regions of the lung compared with normoxic controls. Panel A is from control mice, and panel B is from mice exposed to acute hypoxia. Panel C shows the quantification of ZBTB46⁺ cells around vessels after 3D of hypoxia exposure. p value * =.01 to .05. Vessel integrated intensity: Sum of the pixel intensity divided by all of the pixels in the image (Metamorph^®).

FIGURE 4

cDCs contribute to initiation and maintenance of experimental HPH. (A) Generation of chimera mice and experiment design. (A1) RVSP (right ventricle systolic pressure) of Zbtb46 DTR chimera mice in normoxic and hypoxic conditions (7 days). DT and vehicle injections started one day prior to hypoxia exposure. After cDC depletion, RVSP returned to normoxic levels. (A2) Fulton index of the same mice shows no significant changes but a decreasing trend in right ventricle hypertrophy (RVH). (B) Generation of chimera mice and experiment design. (B1) Zbtb46 DTR chimera mice were protected from PH when cDC depletion occurred after HPH was established (14D hypoxia) and again challenged with hypoxia for 7 days. (B2) Fulton index of mice from panel C shows no significant changes in RVH (Fulton Index). p value * =.01 to .05, ** =.001 to .01, *** =.0001 to .001, **** <.0001, ns, not significant.

FIGURE 5

CD301b DTR chimera (cDC2 depleted), not Batf3^−/− (cDC1 depleted) mice are protected from hypoxia-induced pulmonary hypertension. (A) Experimental design and chimera generation. (A1) RVSPs (right ventricle systolic pressure) were significantly attenuated in CD301b knockout mice (Mgl^−/−), after cDC2 depletion (DT-treated), whereas Fulton Index (right ventricle hypertrophy- RVH) tended to decrease (A2). (B) experiment design. (B1) Batf3 ^−/− mice presented similar RVSP and Fulton Index compared with wild-type controls on day 7 and tended to increase RVSPs after 21 days of hypoxia (B2). NMX (normoxia (Denver altitude), HYX (hypoxia). p value * =.01 to .05, *** =.0001 to .001, **** <.0001, ns, not significant.

FIGURE 6

Interstitial macrophages (IM) are decreased (MHCII⁺, FolR2⁺, CCR2⁺, or CCR2⁻) in DT-treated ZBTB46 DTR chimeras compared with vehicle-treated controls. Panel A shows the gating strategy to isolate interstitial macrophages. Initially, singlets were isolated, followed by exclusion of dead cells, as well as T cells, B cells, and neutrophils (VD + dump gate). CD45⁺ parenchymal cells (IV-CD45 negative), siglec-F negative cells were then gated and CD64⁺ cells were identified as interstitial macrophages. Panel B shows MHCII⁺, FolR2⁺, CCR2⁺ in purple and MHCII⁺, FolR2⁺, CCR2⁻ in red (both from CD64⁺ gate) showing a smaller number of FolR2⁺ and CCR2⁺ or CCR2⁻ in cDCs depleted mice (DT treated). On the right, CCR2 expression in both populations in relation to a fluorescence minus one (FMO) control (dashed blue line) is shown, confirming CCR2 gating. Panel C shows the quantification of both IM populations; Panel D shows that M1-like macrophage-related inflammation was dampened in DT treated mice (cDC depleted), as indicated by a lower expression of iNOS in lung lysates. No differences in M2-like inflammation (Arg-1). p value * =.01 to .05, ** =.001 to .01, ns, not significant.

FIGURE 7

Vascular macrophages are decreased after cDC depletion (DT treated) in ZBTB46 DTR chimera mice challenged with 7 days of hypoxia. Panel A, ZBTB46 DTR chimera Vehicle treated; Panel B, DT treated. Arrows point to CD68⁺ cells. Panel C shows the quantification of macrophages in/ around the vessel. Total vascular CD68⁺ cells are decreased in DT-treated ZBTB46 chimeras compared with PBS-treated controls after 3 days of hypoxia exposure. p value * = .01 to .05. Images magnification: 60 μm.

FIGURE 8

Bulk RNA-Seq data shows a different expression profile after depletion of cDCs after hypoxia exposure. (A) DT and vehicle-treated lungs of ZBTB46 DTR chimera mice (7 days hypoxia) bulk-RNA Seq data cluster into 2 distinct groups in a PCA plot (Red = controls and blue = DT treated). (B) Genes downregulated after cDC depletion using bulk RNA Seq, their fold change in relation to PBS treated controls and adjusted p value.

FIGURE 9

Wisp-1 is a potential mechanism for how cDCs are deleterious in the context of HPH. (A) Western blot showing Wisp-1 expression decreased in cDCs depleted ZBTB46 DTR chimera mice (DT treated). Lanes 1, 2, and 11 are protein markers (Kda). Lanes 3 to 6 are lung lysates from different ZBTB46 DTR chimera mice treated with vehicle, whereas lanes 7 to 10 are lung lysates of different ZBTB46 DTR chimera mice treated with DT (cDC depleted). (B) Western blot quantification confirms that Wisp-1 is decreased after cDCs depletion. β-actin band has a molecular weight of 40 kDa, whereas Wisp-1 has a molecular weight of ~50 kDa.

FIGURE 10

Proteomics assay shows that proteins involved in inflammation are decreased after cDC depletion in mice exposed to hypoxia. (A) Log₂ fold change of protein markers in lungs of DT-treated in relation to PBS treated controls ZBTB46 DTR chimera mice (7 days hypoxia) (Mouse XL Cytokine Array) shows classical cytokines involved in inflammation downregulated after cDC depletion. N = 2/group. (B) Pathway analysis of proteins downregulated after cDC depletion using the Hallmark 2020 database shows inflammation as the main pathway downregulated after cDC depletion (enriched terms on the right of the figure).

FIGURE 11

cDC2s, a classical cDC subset, are increased in the perivascular compartment of patients with IPAH (Idiopathic arterial pulmonary hypertension). Panel A shows cDC2s (in brown) in a control subject; panel B shows cDC2s accumulated around a plexiform lesion of a patient with IPAH (PL: Plexiform lesion; TLO: Tertiary lymphoid organ). Panel C shows cDC2s accumulated around a remodeled vessel of a patient with Schistosoma-induced PAH. Panel D shows the quantification of cDC2s around the vessels. *p value between .01 and .05). Vessel integrated intensity: Sum of the pixel intensity divided by all of the pixels in the image (Metamorph^®).

FIGURE 12

Deconvolution data of vascular lesions in control subjects and IPAH patients (frequencies) shows cDC2s as one of the main immune cell populations, increased mostly in the adventitia and plexiform regions of IPAH patients. No differences were observed in cDC1s. Control adventitia (Ctrl/Adv); Control intima⁺ media (Ctrl/IM); IPAH adventitia (IPAH/Adv.); IPAH obliterative (IPAH/ Oblit.); IPAH intima⁺ media hypertrophy (IPAH/IMH); IPAH plexiform (IPAH/PLX). EC All = all endothelial cell types.

FIGURE 13

HIF1a and CD1a (a marker for cDC2s dendritic cells) double immunohistochemical localization in a control (A, B) or IPAH lungs (C, D). There is minimal expression of HIF1a (dark green, arrows) and CD1a in control pulmonary arteries. IPAH concentric lesions (C) and plexiform lesions (D) show scattered positivity for vascular cells positive for HIFa in the nuclei (arrows), with several dendritic cells (arrowheads) dispersed largely in the adventitia. There is no evidence of HIF1a expression in dendritic, cDC2s, cells (Magnification bar, A–C = 50 μm; D: 100 μm).